Xeon ES/QS Benchmark Database

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

BenchZowner

Member
Nov 16, 2016
36
5
8
38
Help us build a performance database on Intel Xeon ES/QS CPUs ( Single, Dual, Quad processor systems all accepted ).

All you have to do is run the following benchmarks, and post your scores/results along with your CPU & system info.

The benchmarks ( click on their name to download! ) to be used for the database are:

Maxon CineBench R15.038 CPU ( All Cores/Threads ) & CPU Single ( Single Thread ) tests
HWbot.ORG's x265 Benchmark @ 4k Preset 64bit Normal Priority *try with overkill off, then 2x, 3x, 4x, 5x to find where your system nets the best result*
Fritz Chess Benchmark v4.3 with the "Processors to use" set to 4

https://hwbotdownloads.s3.amazonaws.com/downloads/benchmarks/HWBOT_X265_2.0.rar
To run the Single & All CPU tests in CineBench you need to go to File => Advanced Benchmark
and then you get two Run buttons next to the CPU and CPU (Single Core) options.

For x265 for 14c/28t CPUs the sweetspot seems to be 3x overkill.
For higher core count CPUs like 18, 20, and 22 core CPUs you should try 4x and 5x definitely.

If you all agree we can add more benchmarks to the "suite"
 
Last edited:
  • Like
Reactions: Riok and gustav9797

BenchZowner

Member
Nov 16, 2016
36
5
8
38
Let's start with a 2680 v4 E.S. single CPU configuration ( dual CPU should be added soon once I get some time off work )







Edit: for some reason this CPU/board/BIOS combination doesn't boost the active cores properly.
Despite running a Single Core load on CineBench it didn't boost over 23x ( which is the all cores max turbo on this chip ).
I had to manually disable the cores in the BIOS to get the single core max turbo of 26x to work.

The CPU Single Core CineBench R15 score for this CPU is: 107
 
Last edited:

BenchZowner

Member
Nov 16, 2016
36
5
8
38
Why don't we also add Cinebench R11.5 to the benchmark suite. That used to be the standard until R15 came along.
I see no problem except that it would be redundant.
If we are to spend more time benching, we should include something that covers a different application/category than add a second Cinema 4D engine benchmark.
 

gtz

Member
Aug 23, 2016
37
8
8
34
Here are my results.
2683 V3 QS

1832 Multi 127 Single


18.88 9061


Was unable to run the final test. HPET is disabled.
 
Last edited:
  • Like
Reactions: BenchZowner

BenchZowner

Member
Nov 16, 2016
36
5
8
38
follow the instructions provided to enable HPET ( can be disabled afterwards, although it is better left enabled ):

Open a Command Prompt window ( CMD ) with Administrative privileges ( find Command Prompt in the Start Menu and right click on it, select more -> Run as administrator

then type

bcdedit /set useplatformclock true
reboot and run the x265 test ;)
 

gtz

Member
Aug 23, 2016
37
8
8
34
I will re submit my scores now that I have done the all cores max turbo exploit the Haswell Xeons have. Did a quick and dirty run of Cinebench and scored around 21000, after tweaking should be at 22000 with my 2683 V3.
 

Riok

New Member
Dec 28, 2017
10
2
3
47
Hello,

I got a Xeon 1650v3 QEZ0 engineering sample. QDF is QEZ0 (ES2), Stepping is L0. It runs at 3.2Ghz base and turbo at 3.5Ghz.

Performance with 2133mhz. ram, 3.2Ghz:
Cinebench R15 multi score is 968 points.
Firestrike is 10628 points with a GTX 1060.

Performance with 2133mhz. ram, 3.5Ghz:
Cinebench R15 multi score is 1033 points.